setwd("~/Desktop/Datasets")
require(dplyr) 
require(plotrix)
require(lfe)
library(foreign)
library(knitr)
library(car)
library(sandwich)
library(lmtest)
library(plm)
library(Matching)
library(ebal)
library(sandwich)
library(lmtest)
library(lfe)
library(AER)
library(ivpack)
library(stargazer)
library(broom)
library(dotwhisker)
library(foreign)
library(MASS)
library(nnet)
library(dplyr)
library(boot)
library(sampling)

################################################ 
################################################ 
#### Cleaning Data  ############################ 
################################################ 
################################################

capital.data <- read.csv("capital.habeas.judges.readable.no.scotus.csv")
capital.data$Sentence.at.Appeal <- ifelse(capital.data$Sentence.at.Appeal=="Death", 1, 0)
capital.data$Conviction.original <- capital.data$Conviction
capital.data$Conviction <- ifelse((capital.data$Conviction==1|capital.data$Conviction==.5), 1, 0)
capital.data$LC.Conviction <- ifelse((capital.data$LC.Conviction==1|capital.data$LC.Conviction==.5), 1, 0)
capital.data$LC.Sentence <- ifelse((capital.data$LC.Sentence==1|capital.data$LC.Sentence==.5), 1, 0)
capital.data$Type.of.Crime.original <- capital.data$Type.of.Crime
capital.data$Type.of.Crime <- ifelse(capital.data$Type.of.Crime=="Murder 2"|capital.data$Type.of.Crime=="Felony Murder", 0, 1)
capital.data$D.s.Gender <- ifelse(capital.data$D.s.Gender=="M", 1, 0)
capital.data$Jurisidiction <- ifelse(capital.data$Jurisidiction=="Federal", 1, 0)
capital.data$Split.Opinion <- ifelse(capital.data$Split.Opinion=="Yes", 1, 0)
capital.data$race.original <- capital.data$D.s.Race

capital.data$Sentence <- ifelse(capital.data$Sentence=="1"|capital.data$Sentence==".5", 1, 0)
sum(ifelse(capital.data$Sentence.at.Appeal==1, capital.data$Conviction, 0), na.rm = T)

sum(ifelse(capital.data$Sentence.at.Appeal==0, capital.data$Conviction, 0), na.rm = T)

sum(ifelse(capital.data$Sentence.at.Appeal==1 & (capital.data$Conviction==0|capital.data$Conviction==1), 1, 0), na.rm = T)
sum(ifelse(capital.data$Sentence.at.Appeal==0 & (capital.data$Conviction==0|capital.data$Conviction==1), 1, 0), na.rm = T)

###################################################################################
###################################################################################
###################################################################################################
#### Circuit Fixed Effects (From Earlier Coding and For MVN Simulation) ##########################################################################################################################################################################################################
#################################################################

capital.data$Federal.Circuit.Court.Number.1 <- ifelse(capital.data$Federal.Circuit.Court.Number==1, 1, 0)
capital.data$Federal.Circuit.Court.Number.2 <- ifelse(capital.data$Federal.Circuit.Court.Number==2, 1, 0)
capital.data$Federal.Circuit.Court.Number.3 <- ifelse(capital.data$Federal.Circuit.Court.Number==3, 1, 0)
capital.data$Federal.Circuit.Court.Number.4 <- ifelse(capital.data$Federal.Circuit.Court.Number==4, 1, 0)
capital.data$Federal.Circuit.Court.Number.5 <- ifelse(capital.data$Federal.Circuit.Court.Number==5, 1, 0)
capital.data$Federal.Circuit.Court.Number.6 <- ifelse(capital.data$Federal.Circuit.Court.Number==6, 1, 0)
capital.data$Federal.Circuit.Court.Number.7 <- ifelse(capital.data$Federal.Circuit.Court.Number==7, 1, 0)
capital.data$Federal.Circuit.Court.Number.8 <- ifelse(capital.data$Federal.Circuit.Court.Number==8, 1, 0)
capital.data$Federal.Circuit.Court.Number.9 <- ifelse(capital.data$Federal.Circuit.Court.Number==9, 1, 0)
capital.data$Federal.Circuit.Court.Number.10 <- ifelse(capital.data$Federal.Circuit.Court.Number==10, 1, 0)
capital.data$Federal.Circuit.Court.Number.11 <- ifelse(capital.data$Federal.Circuit.Court.Number==11, 1, 0)
capital.data$Federal.Circuit.Court.Number.D.C. <- ifelse(capital.data$Federal.Circuit.Court.Number=="D.C.", 1, 0)
capital.data$Federal.Circuit.Court.Number.SCOTUS <- ifelse(capital.data$Federal.Circuit.Court.Number=="SCOTUS", 1, 0)

####################################################################################
##################################################################################
#### Subsetting Cases for Analysis of Potential Trump Appointees ##################
##################################################################################
####################################################################################

Sykes.cases <- capital.data[capital.data$Judge=="Sykes",]
Tymkovich.cases <- capital.data[capital.data$Judge=="Tymkovich",]
Colloton.cases <- capital.data[capital.data$Judge=="Colloton",]
Hardiman.cases <- capital.data[capital.data$Judge=="Hardiman",]
Willet.cases <- capital.data[capital.data$Judge=="Willet",]
Kethledge.cases <- capital.data[capital.data$Judge=="Kethledge",]
Barrett.cases <- capital.data[capital.data$Judge=="Barrett",]
Gruender.cases <- capital.data[capital.data$Judge=="Gruender",]
Stras.cases <- capital.data[capital.data$Judge=="Stras",]
Eid.cases <- capital.data[capital.data$Judge=="Eid",]
Gorsuch.cases <- capital.data[capital.data$Judge=="Gorsuch",]
Grant.cases <- capital.data[capital.data$Judge=="Grant",]
Newsom.cases <- capital.data[capital.data$Judge=="Newsom",]
Pryor.cases <- capital.data[capital.data$Judge=="Pryor",]
Thapar.cases <- capital.data[capital.data$Judge=="Thapar",]
Larsen.cases <- capital.data[capital.data$Judge=="Larsen",]
Kavanaugh.cases <- capital.data[capital.data$Judge=="Kavanaugh",]

############################
############################
############################
#### More Cleaning Data  #### 
############################
############################
############################

capital.data <- capital.data[capital.data$D.s.Race!="Unknown",]

#### For testing more specific associations between race and outcome

capital.data$D.s.Race <- ifelse(capital.data$D.s.Race=="White", 1, 0)

### Coding Judge Party

capital.data$Party <- ifelse((capital.data$Judge=="Reinhardt"|capital.data$Judge=="Pregerson"|capital.data$Judge=="Nyguen"|capital.data$Judge=="Farris"|capital.data$Judge=="Owens"|capital.data$Judge=="Restrepo"|capital.data$Judge=="Fisher"|capital.data$Judge=="Phillips"|capital.data$Judge=="McHugh"|capital.data$Judge=="A. Jordan"|capital.data$Judge=="R. Anderson"|capital.data$Judge=="Hamilton"|capital.data$Judge=="Wood"|capital.data$Judge=="A. Williams"|capital.data$Judge=="Wardlaw"|capital.data$Judge=="Daughtrey"|capital.data$Judge=="Donald"|capital.data$Judge=="Keith"|capital.data$Judge=="Gregory"|capital.data$Judge=="Wynn"|capital.data$Judge=="Diaz"|capital.data$Judge=="Hull"|capital.data$Judge=="Marcus"|capital.data$Judge=="King"|capital.data$Judge=="Matheson"|capital.data$Judge=="J. Carnes"|capital.data$Judge=="Lucero"|capital.data$Judge=="Ginsburg"|capital.data$Judge=="Breyer"|capital.data$Judge=="Sotomayor"|capital.data$Judge=="Kagan"|capital.data$Judge=="Gould"|capital.data$Judge=="Shwartz"|capital.data$Judge=="Krause"|capital.data$Judge=="Clay"|capital.data$Judge=="Hood"|capital.data$Judge=="Martin"|capital.data$Judge=="J. Pryor"|capital.data$Judge=="Tallman"|capital.data$Judge=="Wilson"|capital.data$Judge=="Rosenbaum"|capital.data$Judge=="Ambro"|capital.data$Judge=="Greenaway"|capital.data$Judge=="Scirica"|capital.data$Judge=="Christen"|capital.data$Judge=="Nguyen"|capital.data$Judge=="Watford"|capital.data$Judge=="Rawlinson"|capital.data$Judge=="Vanaskie"|capital.data$Judge=="Kayatta"|capital.data$Judge=="Barron"|capital.data$Judge=="J. Kelly"|capital.data$Judge=="McKeown"|capital.data$Judge=="Stewart"|capital.data$Judge=="McKee"|capital.data$Judge=="Fuentes"|capital.data$Judge=="Silverman"|capital.data$Judge=="Moore"|capital.data$Judge=="Cole"|capital.data$Judge=="Schroeder"|capital.data$Judge=="Murguia"|capital.data$Judge=="Dennis"|capital.data$Judge=="Graves"|capital.data$Judge=="Fletcher"|capital.data$Judge=="Friedland"|capital.data$Judge=="Gilman"|capital.data$Judge=="Gilman"|capital.data$Judge=="Berzon"|capital.data$Judge=="Briscoe"|capital.data$Judge=="Motz"|capital.data$Judge=="DeGuilio"|capital.data$Judge=="Hornak"|capital.data$Judge=="Floyd"|capital.data$Judge=="Molloy"|capital.data$Judge=="Mendoza"|capital.data$Judge=="Schreier"|capital.data$Judge=="Higginson"|capital.data$Judge=="Merritt"|capital.data$Judge=="Traxler"|capital.data$Judge=="Thacker"|capital.data$Judge=="Thompson"|capital.data$Judge=="Wilkins"|capital.data$Judge=="Harris"|capital.data$Judge=="Pooler"|capital.data$Judge=="Lohier"|capital.data$Judge=="Rendell"|capital.data$Judge=="Bastian"|capital.data$Judge=="Stranch"|capital.data$Judge=="Graber"|capital.data$Judge=="Bennett"|capital.data$Judge=="Pearson"|capital.data$Judge=="Lynch"|capital.data$Judge=="Paez"|capital.data$Judge=="Reavley"|capital.data$Judge=="Beckwith"|capital.data$Judge=="Marmolejo"|capital.data$Judge=="Teilborg"|capital.data$Judge=="Hull"|capital.data$Judge=="Murphy"|capital.data$Judge=="Moritz"|capital.data$Judge=="Lipez"|capital.data$Judge=="Bacharach"|capital.data$Judge=="Hawkins"|capital.data$Judge=="Hurwitz"|capital.data$Judge=="Collins"|capital.data$Judge=="Gwin"|capital.data$Judge=="Sloviter"|capital.data$Judge=="Daughtry"|capital.data$Judge=="Leitman"|capital.data$Judge=="Calabresi"|capital.data$Judge=="Nelson"|capital.data$Judge=="Huck"|capital.data$Judge=="Adelman"|capital.data$Judge=="Bye"|capital.data$Judge=="Aldisert"|capital.data$Judge=="Timlin"|capital.data$Judge=="Sargus"|capital.data$Judge=="Hellerstein"|capital.data$Judge=="Barry"|capital.data$Judge=="Rice"|capital.data$Judge=="Collier"|capital.data$Judge=="Helmick"|capital.data$Judge=="Marbley"|capital.data$Judge=="Barkett"|capital.data$Judge=="Lemelle"|capital.data$Judge=="Canby"|capital.data$Judge=="Tashima"|capital.data$Judge=="O'Malley"|capital.data$Judge=="Cabranes"|capital.data$Judge=="Steeh"|capital.data$Judge=="Cudahy"|capital.data$Judge=="Carr"), 1, 0)

capital.data$race.party.interaction <- capital.data$D.s.Race*capital.data$Party

######################################
######################################
##############################################################################
##############################################################################
################################################################################
#### Table A.3 Regressions ################
################################################################################
##############################################################################
##############################################################################
##############################################################################
##############################################################################

## Column 1

OLS.regression <- felm(Conviction ~  Sentence.at.Appeal +  D.s.Gender +  D.s.Race +  Party + Type.of.Crime +  Number.of.Victims +  Jurisidiction  + race.party.interaction + Year|Federal.Circuit.Court.Number + State.of.Imprisonment|0|0, data = capital.data)
summary(OLS.regression)

## Column 2

OLS.regression.judge.fixed <- felm(Conviction ~  Sentence.at.Appeal +  D.s.Gender +  D.s.Race +  Type.of.Crime +  Number.of.Victims +  Jurisidiction + Year|Judge + State.of.Imprisonment|0|0, data = capital.data)
summary(OLS.regression.judge.fixed)

## Column 3
CSE.regression <- felm(Conviction ~  Sentence.at.Appeal +  D.s.Gender +  D.s.Race +  Party + Type.of.Crime +  Number.of.Victims +  Jurisidiction  + race.party.interaction + Year|Federal.Circuit.Court.Number + State.of.Imprisonment|0|Case.Name, data = capital.data)
summary(CSE.regression)

## Column 4

CSE.regression.judge.fixed <- felm(Conviction ~ Sentence.at.Appeal +  D.s.Gender +  D.s.Race + Type.of.Crime +  Number.of.Victims +  Jurisidiction + Year|Judge + State.of.Imprisonment|0|Case.Name, data = capital.data)
summary(CSE.regression.judge.fixed)

## Column 5

check.1 <- glm(Conviction ~  Sentence.at.Appeal +  D.s.Gender +  D.s.Race +  Party + Type.of.Crime +  Number.of.Victims +  Jurisidiction  + race.party.interaction + Year + as.factor(Federal.Circuit.Court.Number), data=capital.data, family=binomial(link='logit'))

logit.model.cse.court.fixed <- coeftest(check.1, vcov = vcovHC(check.1, type = "HC1", cluster = capital.data$Case.Name))

## Column 6

check.2 <- glm(Conviction ~  Sentence.at.Appeal +  D.s.Gender +  D.s.Race + Type.of.Crime +  Number.of.Victims +  Jurisidiction + Year + as.factor(Judge), data=capital.data, family=binomial(link='logit'))

logit.model.cse.judge.fixed <- coeftest(check.2, vcov = vcovHC(check.2, type = "HC1", cluster = capital.data$Case.Name))


### Producing the Latex Code for Table A.3

covariate.names.nv <- c("Sentence at Time of Appeal (1 if Death, 0 Otherwise)", "Male (1 if Male, 0 Otherwise)", "Race (1 if White-Non-Hispanic, 0 Otherwise)", "Party of Appointing President (1 if Dem, 0 Otherwise)", "Type of Crime (1 if 1st Degree Murder or Equivalent, 0 Otherwise)", "Number of Victims", "Jurisdiction (1 if Federal, 0 Otherwise)", "Interaction of Race and Party", "Year")
dv.name.nv <- c("Guilt-Phase Vote (1 if Favorable to Defendant, 0 Otherwise)")

require("stargazer")
stargazer(OLS.regression, OLS.regression.judge.fixed, CSE.regression, CSE.regression.judge.fixed, logit.model.cse.court.fixed, logit.model.cse.judge.fixed, title="Table A.3: Effects of Death Sentence on Probability of Favorable Habeas Ruling (No SCOTUS)", align=T, type = "latex", report = ('vc*s'), multicolumn=T, covariate.labels = covariate.names.nv, omit=c("Federal.Circuit.Court.Number", "Judge", "State.of.Imprisonment", "Constant"), dep.var.labels = dv.name.nv, column.labels = c("OLS", "OLS", "CSE", "CSE", "Logit CSE", "Logit CSE"), column.separate = c(1,1), add.lines = list(c("Court-Level Fixed Effects", "Yes","No","Yes","No", "Yes","No"), c("Judge-Level Fixed Effects", "No","Yes","No","Yes","No","Yes"), c("State-Level Fixed Effects", "Yes","Yes","Yes","Yes","No","No"), c("Observations", "1,339","1,339","1,339","1,339","1,339","1,339")), omit.stat = c("adj.rsq", "f", "ser", "rsq", "n"), model.names = F, notes = "Interpretations: *p < .1; **p < .05; ***p < .01 ", notes.append = F, notes.align = "l", column.sep.width = "-15pt")

################################################################
################################################################
##### Party Seperation (Table 3) ##########################################
##############################################################
################################################################

## Subsetting to Democratic and Republican Judges

capital.data.democratic.judges <- capital.data[capital.data$Party==1,]
capital.data.republican.judges <- capital.data[capital.data$Party==0,]

# Column 1

OLS.regression.democratic.judges <- felm(Conviction ~  Sentence.at.Appeal +  D.s.Gender +  D.s.Race +  Type.of.Crime +  Number.of.Victims +  Jurisidiction + Year|Federal.Circuit.Court.Number + State.of.Imprisonment|0|0, data = capital.data.democratic.judges)

summary(OLS.regression.democratic.judges)

# Column 2

CSE.regression.democratic.judges <- felm(Conviction ~  Sentence.at.Appeal +  D.s.Gender +  D.s.Race +  Type.of.Crime +  Number.of.Victims +  Jurisidiction + Year|Federal.Circuit.Court.Number + State.of.Imprisonment|0|Case.Name, data = capital.data.democratic.judges)

summary(CSE.regression.democratic.judges)

# Column 3

check.3 <- glm(Conviction ~  Sentence.at.Appeal +  D.s.Gender +  D.s.Race +  Type.of.Crime +  Number.of.Victims +  Jurisidiction + Year + as.factor(Federal.Circuit.Court.Number), data=capital.data.democratic.judges, family=binomial(link='logit'))

logit.model.dem <- coeftest(check.3, vcov = vcovHC(check.3, type = "HC1", cluster = capital.data.democratic.judges$Case.Name))

# Column 4

OLS.regression.rep.judges <- felm(Conviction ~  Sentence.at.Appeal +  D.s.Gender +  D.s.Race +  Type.of.Crime +  Number.of.Victims +  Jurisidiction + Year|Federal.Circuit.Court.Number + State.of.Imprisonment|0|0, data = capital.data.republican.judges)
summary(OLS.regression.rep.judges)

# Column 5

CSE.regression.rep.judges <- felm(Conviction ~  Sentence.at.Appeal +  D.s.Gender +  D.s.Race +  Type.of.Crime +  Number.of.Victims +  Jurisidiction + Year|Federal.Circuit.Court.Number + State.of.Imprisonment|0|Case.Name, data = capital.data.republican.judges)

summary(CSE.regression.rep.judges)

# Column 6

check.4 <- glm(Conviction ~  Sentence.at.Appeal +  D.s.Gender +  D.s.Race +  Type.of.Crime +  Number.of.Victims +  Jurisidiction + Year + as.factor(Federal.Circuit.Court.Number), data=capital.data.republican.judges, family=binomial(link='logit'))

logit.model.rep <- coeftest(check.4, vcov = vcovHC(check.4, type = "HC1", cluster = capital.data.republican.judges$Case.Name))



### Latex Code for Table 3 ###

covariate.names.nv.2 <- c("Sentence at Time of Appeal (1 if Death, 0 Otherwise)", "Male (1 if Male, 0 Otherwise)", "Race (1 if White-Non-Hispanic, 0 Otherwise)", "Type of Crime (1 if 1st Degree Murder or Equivalent, 0 Otherwise)", "Number of Victims", "Jurisdiction (1 if Federal, 0 Otherwise)", "Year")
dv.name.nv.2 <- c("Guilt-Phase Vote", "Guilt-Phase Vote")

require("stargazer")

stargazer(OLS.regression.democratic.judges, CSE.regression.democratic.judges, logit.model.dem, OLS.regression.rep.judges, CSE.regression.rep.judges, logit.model.rep, title="Table A.4: Effects of Death Sentence on Probability of Favorable Habeas Ruling—Effects Seperated By Party (No SCOTUS)", align=TRUE, type = "latex", report = ('vc*s'), multicolumn=T, covariate.labels = covariate.names.nv.2, omit=c("Federal.Circuit.Court.Number", "State.of.Imprisonment", "Constant"), dep.var.labels = c(" ", "Guilt-Phase Vote", " ", " ", "Guilt-Phase Vote", " "), column.labels = c("Democratic-Appointed Judges", "Republican-Appointed Judges"), column.separate = c(3,3), add.lines = list(c("Court-Level Fixed Effects", "Yes","Yes", "Yes", "Yes", "Yes", "Yes"), c("State-Level Fixed Effects", "Yes","Yes", "No", "Yes", "Yes", "No"), c("Standard Errors", "OLS", "CSE", "Logit", "OLS", "CSE", "Logit"), c("Observations", "674", "674", "674", "665", "665", "665")), omit.stat = c("adj.rsq", "f", "ser", "n", "rsq"), model.names = F, column.sep.width = "-15pt", notes = "Interpretations: *p < .1; **p < .05; ***p < .01 ", notes.append = F, notes.align = "l")

## General Rates of Relief Under Democratic and Republican Judges

mean(capital.data.democratic.judges$Conviction, na.rm = T)
mean(capital.data.republican.judges$Conviction, na.rm = T)

###########################################################################
######################################################################################
####### First Degree Murder or Equivalent Only (Table 4) ##############################
#####################################################################################
###########################################################################


## Subsetting to First Degree Murder Cases

capital.data.fdm <- capital.data[capital.data$Type.of.Crime==1,]

## Column 1

OLS.regression.fdm <- felm(Conviction ~  Sentence.at.Appeal +  D.s.Gender +  D.s.Race +  Party + Number.of.Victims +  Jurisidiction  + race.party.interaction + Year|Federal.Circuit.Court.Number + State.of.Imprisonment|0|0, data = capital.data.fdm)
summary(OLS.regression.fdm)

## Column 2

OLS.regression.fdm.judge.fixed <- felm(Conviction ~  Sentence.at.Appeal +  D.s.Gender +  D.s.Race + Number.of.Victims +  Jurisidiction  + Year|Judge + State.of.Imprisonment|0|0, data = capital.data.fdm)
summary(OLS.regression.fdm.judge.fixed)

## Column 3

CSE.regression.fdm <- felm(Conviction ~  Sentence.at.Appeal +  D.s.Gender +  D.s.Race +  Party + Number.of.Victims +  Jurisidiction  + race.party.interaction + Year|Federal.Circuit.Court.Number + State.of.Imprisonment|0|Case.Name, data = capital.data.fdm)
summary(CSE.regression.fdm)

## Column 4

CSE.regression.fdm.judge.fixed <- felm(Conviction ~  Sentence.at.Appeal +  D.s.Gender +  D.s.Race + Number.of.Victims +  Jurisidiction  + Year|Judge + State.of.Imprisonment|0|Case.Name, data = capital.data.fdm)
summary(CSE.regression.judge.fixed)

## Column 5

check.5 <- glm(Conviction ~  Sentence.at.Appeal +  D.s.Gender +  D.s.Race +  Party + Number.of.Victims +  Jurisidiction  + race.party.interaction + Year + as.factor(Federal.Circuit.Court.Number), data=capital.data.fdm, family=binomial(link='logit'))

logit.model.cse.court.fixed.fdm <- coeftest(check.5, vcov = vcovHC(check.5, type = "HC1", cluster = capital.data.fdm$Case.Name))

## Observing Problem of Perfect/Near Perfect Seperation in footnote 113

check.6 <- glm(Conviction ~  Sentence.at.Appeal +  D.s.Gender +  D.s.Race +  Party + Number.of.Victims +  Jurisidiction  + race.party.interaction + Year + as.factor(Judge), data=capital.data.fdm, family=binomial(link='logit'))

### Producing the Latex Code for Table A.5

covariate.names.nv.3 <- c("Sentence at Time of Appeal (1 if Death, 0 Otherwise)", "Male (1 if Male, 0 Otherwise)", "Race (1 if White-Non-Hispanic, 0 Otherwise)", "Party of Appointing President (1 if Dem, 0 Otherwise)", "Number of Victims", "Jurisdiction (1 if Federal, 0 Otherwise)", "Interaction of Race and Party", "Year")
dv.name.nv <- c("Guilt-Phase Vote (1 if Favorable to Defendant, 0 Otherwise)")

require("stargazer")
stargazer(OLS.regression.fdm, OLS.regression.fdm.judge.fixed, CSE.regression.fdm, CSE.regression.fdm.judge.fixed, logit.model.cse.court.fixed.fdm, title="Table A.5: Effects of Death Sentence on Probability of Favorable Habeas Ruling — First Degree Murder and Equivalents Only (No SCOTUS)", align=TRUE, type = "latex", report = ('vc*s'), multicolumn=T, covariate.labels = covariate.names.nv.3, omit=c("Constant", "Federal.Circuit.Court.Number", "Judge", "State.of.Imprisonment"), dep.var.labels = dv.name.nv, column.labels = c("OLS","CSE", "Logit"), column.separate = c(2,2,1), add.lines = list(c("Court-Level Fixed Effects", "Yes","No","Yes","No", "Yes"), c("Judge-Level Fixed Effects", "No","Yes","No","Yes","No"), c("State-Level Fixed Effects", "Yes","Yes","Yes","Yes", "No"), c("Observations", "1,193", "1,193", "1,193", "1,193", "1,193") ), omit.stat = c("adj.rsq", "f", "ser", "rsq", "n"), model.names = F, notes = "Interpretations: *p < .1; **p < .05; ***p < .01 ", notes.append = F, notes.align = "l")

